- Boost-users - lists.preview.boost.org

Re: [Boost-users] A question on overloaded function? Can anyone with kindness help me?
by fmingu 30 Jun '09

30 Jun '09

Oh,I made a mistake. And can I use: bind(&std::pow,constant(-1),bind(&IIPrimemap::size,var(primemapvec))); in the program? where : typedef std::map<int, int> IIPrimemap; IIPrimemap primemapvec; the last two lines are corrected.

1 0

[IOSTREAMS] Forcing a flush on newlines
by David Larsson 30 Jun '09

30 Jun '09

Hi! I'm outputting the progress from an externally executed command line application using the windows API call OutputDebugString. This means that I need to convert all line breaking characters to flush commands to make sure the text is outputted when a new line is received rather than when some underlying structure finds it suitable to linebreak. I've created a sink that writes to OutputDebugString and it works great, however unless I manually insert lots of std::endl (or flushing some other way) I won't get any output until the buffer is full. Is it possible to write an iostream filter that forces a flush on newlines or is this something I need to handle before feeding data into my stream? /David

2 1

A question on if_then_else in lambda.Can any one with kindness help me?
by fmingu 30 Jun '09

30 Jun '09

I use Dev-c++, and found a problem: # indcude "boost/lambda/if ............... typedef std::map<int, int> IIPrimemap; IIPrimemap primemapvec; ....................... int nnumber; int switchvalue; ........................ if_then_else(var(nnumber)==1,var(switchvalue)=1, if_then_else(count_if(primemapvec.begin(),primemapvec.end(), bind(&IIPrimemap::value_type::second,_1)>2)>0, var(switchvalue)=2,var(switchvalue)=3)); ........................ But the complier told me that: no matching function for call to `if_then_else(bool, const boost::lambda::lambda_functor<boost::lambda::lambda_functor_base<boost::lambda::other_action<boost::lambda::assignment_action>, I do not know how to handle it. Can any one with knidness help me?

2 1

test
by Sergey Skorokhodov 29 Jun '09

29 Jun '09

sorry

1 0

[lambda] shared_ptr's in lambda expressions
by Michael Crawford 29 Jun '09

29 Jun '09

I have the following piece of code: void MultiLog::removeLog(const std::string& name) { std::remove_if(m_logs.begin( ), m_logs.end( ), (boost::lambda::_1 ->* &Log::getName) == name); } Which fails to compile with the typical spectacular list of template error messages, the first of which is "error C2784 ... could not deduce template argument for 'const boost::lambda::lambda_functor<T> &' from 'boost::lambda::detail::member_pointer_caller<RET,A,B>'" I'm assuming that the error is because each element here is a shared_ptr<Log>, rather than a raw pointer to Log. Using bind to wrap shared_ptr::get appears to work: void MultiLog::removeLog(const std::string& name) { std::remove_if(m_logs.begin( ), m_logs.end( ), boost::bind(&Log::getName, boost::bind(&LogPtr::get, _1)) == name); } but is there are way to use the significantly cleaner & clearer member pointer operator syntax while using shared_ptr's? Thanks, MBC

2 1

Spirit error when building serialization
by Pascal Kesseli 29 Jun '09

29 Jun '09

Hi again Running bjam with an mxscaleb-gcc cross compilation toolchain (Version 3.4.3), I repeatedly get the following error line on boost spirit: In file included from ./boost/spirit/home/classic/core.hpp:33, from ./boost/spirit/include/classic_core.hpp:11, from ./boost/graph/detail/read_graphviz_spirit.hpp:32, from ./boost/graph/graphviz.hpp:780, from libs/graph/src/read_graphviz_spirit.cpp:28: ./boost/spirit/home/classic/core/primitives/numerics.hpp:124: Error: Declaration »template<class ScannerT> struct boost::spirit::classic::sign_parser::resultÂ« ./boost/spirit/home/classic/core/parser.hpp:125: conflicts with previous declaration of <class ScannerT> struct boost::spirit::classic::parser<boost::spirit::classic::sign_parser>::resultÂ « As a result, building the serialization library fails. Any ideas on how to fix this here? Thanks in advance and greets Kessi

1 0

Re: [Boost-users] Newbie needs yet more help with serialization-BOOST_SERIALIZATION_SPLIT_FREE
by Adam Szeliga 29 Jun '09

29 Jun '09

On 2007-08-20 11:03, Hector C. wrote: You need include file: #include <boost/serialization/split_free.hpp> Regards Adam

1 0

mpl / preprocessor question
by Zachary Turner 29 Jun '09

29 Jun '09

Is there a way, using some combination of boost preprocessor and mpl, that I can write a class template<class T, int n> class foo { public: void foo(T arg1, T arg2, ..., T argn); }; In other words, generate a function that requires n arguments where n is a template parameter?

2 1

boost.fusion compile error using VC 8: error C2039: 'tag_of' : is not a member of 'boost::fusion::traits::detail'
by Tang Jiangjun 29 Jun '09

29 Jun '09

When I compiled the code snippet as follows, there were some errors about searching tag_of in namespace boost::fusion::traits::detail. But after reviewing the source code, I found the desired tag_of was actually located in namespace boost::fusion::detail. Moreover it will be compiled successfully if adding this full and correct namespace before tag_of in the file <is_sequence.hpp>. Here is code snippet: typedef fusion::vector< int, string > TPerson; TPerson person( 20, "Alan" ); cout << person << endl; Here are compile errors: 1>d:\libraries\boost_1_39_0\boost\fusion\support\is_sequence.hpp(59) : error C2039: 'tag_of' : is not a member of 'boost::fusion::traits::detail' 1> d:\libraries\boost_1_39_0\boost\utility\enable_if.hpp(36) : see reference to class template instantiation 'boost::fusion::traits::is_sequence<T>' being compiled 1> with 1> [ 1> T=TPerson 1> ] 1> d:\samples\boost\fusion\fusion.cpp(28) : see reference to class template instantiation 'boost::enable_if<Cond,T>' being compiled 1> with 1> [ 1> Cond=boost::fusion::traits::is_sequence<TPerson>, 1> T=std::ostream & 1> ] Here is the origin code in <is_sequence.hpp>: namespace traits { template <typename T> struct is_sequence : extension::is_sequence_impl<typename detail::tag_of<T>::type>:: template apply<T> {}; } Here is the modification: namespace traits { template <typename T> struct is_sequence : extension::is_sequence_impl<typename boost::fusion::detail::tag_of<T>::type>:: template apply<T> {}; } I guess this should be a minor bug in boost.fusion.

2 2

Threaded Daemon - boost::mutex::lock(): Assertion `!pthread_mutex_lock(&m)' failed.
by John D 28 Jun '09

28 Jun '09

Hi This seems an interesting error. I have an application that when executed as ./execute runs just as fine as can be. However when run with a a daemon that loads on boot, when activated it immediately crashes with. /usr/include/boost/thread/pthread/mutex.hpp:50: void boost::mutex::lock(): Assertion `!pthread_mutex_lock(&m)' failed. The cause of the problem clearly is the lack of an in your face shell which I know and is fine. I am wondering what advice anyone might have for having a stable boost threaded tty less executable that will run. I know it is something very small to change likely in the call to execute the application but please any help is appreciated, Thank you, John D

2 1

[asio] STL issues OS X 10.5 (intel)
by Jeffrey Aigner 28 Jun '09

28 Jun '09

I am trying to learn the asio library via the tutorials on the website. I am using XCode on OS X 10.5 (intel). After compiling successfully, I get runtime errors (after program listing): Program: #include <iostream> #include <boost/asio.hpp> #include <boost/date_time/posix_time/posix_time.hpp> int main() { boost::asio::io_service io; boost::asio::deadline_timer t(io, boost::posix_time::seconds(5)); t.wait(); std::cout << "Hello, world!\n"; return 0; } Runtime Errors: Macintosh:Debug jaigner$ ./proxy /Developer/SDKs/MacOSX10.5.sdk/usr/include/c++/4.0.0/debug/ safe_iterator.h:127: error: attempt to copy-construct an iterator from a singular iterator. Objects involved in the operation: iterator "this" @ 0x0x1002b0 { type = N11__gnu_debug14_Safe_iteratorIN10__gnu_norm14_List_iteratorISt4pairIPvPN5boost4asio6detail11timer_queueINS6_11time_traitsINS5_10posix_time5ptimeEEEE10timer_baseEEEEN15__gnu_debug_def4listISG_SaISG_EEEEE (mutable iterator); state = singular; } iterator "other" @ 0x0xbffff4e4 { type = N11__gnu_debug14_Safe_iteratorIN10__gnu_norm14_List_iteratorISt4pairIPvPN5boost4asio6detail11timer_queueINS6_11time_traitsINS5_10posix_time5ptimeEEEE10timer_baseEEEEN15__gnu_debug_def4listISG_SaISG_EEEEE (mutable iterator); state = singular; } Abort trap Any help would be appreciated! --Jeff Aigner

1 0

A question on for_each() and placeholder.Can any one with kindness help me?
by fmingu 28 Jun '09

28 Jun '09

By using Dev-C++, I used lambda expression, I want to know that if I white: typedef std::vector<int> vect1; std::vector<vect1> vect2; vect2 vect2Instance; then std::for_each(vect2Instance.begin(),vect2Instance.end(), std::for_each(_1.begin(),_1.end(),_1+=2)); is legal or not? And if legal ,Does the first two _1's mean a number in vect2 (i.e. a vect1) and the last _1 is a number in vect1(i.e. a int)? Can any one with kindness help me?

2 1

variant within a recursive_variant?
by Matthew L. Creech 27 Jun '09

27 Jun '09

Hi, I'm wondering if anyone has successfully embedded a boost::variant<> within a recursive variant. The program below illustrates what I'm trying to do: ============= #include <vector> #include <string> #include <boost/variant.hpp> typedef boost::variant< double, std::string > a_t; struct b_t { }; typedef boost::make_recursive_variant< a_t, b_t, std::vector<boost::recursive_variant_> >::type c_t; int main() { a_t a(20.0); c_t c(a); return 0; } ============= With GCC, this gives me: ... boost/variant/variant.hpp:1269: error: no matching function for call to 'boost::detail::variant::initializer_root::initialize(void*, double&)' Normal variant-within-a-variant works fine (i.e. if I remove the vector, and just declare c_t without make_recursive_variant<>). And make_recursive_variant<> with plain types works as well (i.e. if a_t is a non-variant type). But I can't figure out how to make the two work together... Any help would be greatly appreciated. Thanks! -- Matthew L. Creech

1 0

How to Compile Boost for All Versions of Visual Studio
by Altaf Navalur 27 Jun '09

27 Jun '09

Hello, I have installed VS 2003, VS 2005 & VS 2008 in my PC. I have to work on all these versions. When I compile boost, it only compiles for VS 2008. How do I compile it for all the three IDE`s? Thanks, Altaf -- An eye for an eye only ends up making the whole world blind.

7 15

fusion::result_of::invoke_function_object and references
by Mike Tegtmeyer 27 Jun '09

27 Jun '09

I have another question regarding compilation failures with fusion. If this has been discussed before, can someone please point me to the right location. If not... Is there some reason the application of result_of::invoke_function_object fails to compile for sequence argument references? ie: result_of::invoke_function_object<doit, fusion::vector<int>&>::type Digging a little, it seems that the failure is due to the size<> operator doesn't work with reference types (or is expecting the reference to be stripped prior to this point). I understand that to avoid unnecessary copies, tiers are typically used but in the case where chains of invoke_function_object would like to be strung together where there is an assignment or construction from the actual return value to a variable declared as the type obtained from result_of (in a fold for example), it seems that the vector copy is unavoidable (at least in my reading) unless the argument vector is a const reference. Is this an oversight or am I missing something? Basically I am trying to string together N function objects in the sequence where the ith object gets called with the result of the i-1th object (where each has a different argument list) using fold and I am running into extraneous copies of the return value. I can post a full example if needed. Thanks in advance, Mike error message: /opt/local/include/boost/fusion/sequence/intrinsic/size.hpp:28: error: const boost::fusion::vector<int, boost::fusion::void_, boost::fusion::void_, boost::fusion::void_, boost::fusion::void_, boost::fusion::void_, boost::fusion::void_, boost::fusion::void_, boost::fusion::void_, boost::fusion::void_>& is not a class, struct, or union type See code below. #include <boost/fusion/container/vector.hpp> #include <boost/fusion/functional/invocation/invoke_function_object.hpp> namespace b = boost; namespace bf = boost::fusion; struct doit { template <typename Sig> struct result; template <typename Self, typename T> struct result< Self(const T &) > { typedef int type; }; template <typename T> int operator()(const T &t) const { return 1; } }; #if !OK typedef bf::result_of:: invoke_function_object<doit,const bf::vector<int>&>::type result_type2; #else typedef bf::result_of:: invoke_function_object<doit,const bf::vector<int> >::type result_type2; #endif int main() {return 0;}

2 2

Compile errors inside boost::asio code I can't affect.
by Michael.Broida＠thomsonreuters.com 27 Jun '09

27 Jun '09

Hi! We've been using Boost 1.38.0: WinXP SP3 on 32-bit box VisualStudio 2005, MSVC++ 8.0, 32-bit builds. I've just rebuilt Boost 1.38.0 on another system for 64-bit builds: Windows Server 2008 Enterprise SP1 VisualStudio 2008, MSVC++ 9.0, 64-bit builds. Found no visible build errors during the Boost build. Yet when I build our own code on the 64-bit system, I get these three compile errors deep inside boost::asio code that I have no control over: >>---------------- error C2664: 'QueueUserAPC' : cannot convert parameter 1 from 'void (__cdecl *)(ULONG)' to 'PAPCFUNC' D:\dev\3rdParty_x64\include\boost\asio\detail\win_thread.hpp 151 >>---------------- error C2664: 'GetQueuedCompletionStatus' : cannot convert parameter 3 from 'DWORD *' to 'PULONG_PTR' D:\dev\3rdParty_x64\include\boost\asio\detail\win_iocp_io_service.hpp 142 >>---------------- error C2664: 'GetQueuedCompletionStatus' : cannot convert parameter 3 from 'DWORD *' to 'PULONG_PTR' D:\dev\3rdParty_x64\include\boost\asio\detail\win_iocp_io_service.hpp 430 >>---------------- The "D:\dev\3rdParty_x64" is where I've installed the includes & libs for Boost and other components. As you can see, those errors are not in our code; they're deep in the ASIO code itself, and I don't see how we could be causing them. I didn't write our code that uses ASIO, so I don't know anything about what ASIO is trying to do there. :( Since I couldn't find any other complaints online about this, I'm flailing around for support. Any pointers would be appreciated. Pertinent data: 64-bit WindowsServer2008 SP1 on Intel 64-bit box VisualStudio2008 SP1 (SP2 is still RC, I think, so we can't use it.) Our projects using "x64" configuration, so uses x64 compiler. Boost built with x64 compiler via: toolset=msvc-9.0 address-model=64 Thanks! Mike This email was sent to you by Thomson Reuters, the global news and information company. Any views expressed in this message are those of the individual sender, except where the sender specifically states them to be the views of Thomson Reuters.

3 8

[format] Thread safety of boost::format
by James Madison 26 Jun '09

26 Jun '09

I was browsing the boost::format docs and did not find anything about the Thread safety of boost::format. Essentially I want to craft a bunch of error handling methods like this: void ProcessError(...args...) { static boost::format formatter(...); //Try to save the cost of creating this for every single error //do stuff LOG(boost::str(formatter % ...args...); } This method would be called from many, many threads. Is this considered safe? Is there some general information about the thread safety of boost::format methods somewhere in the docs?

3 2

[MultiArray] Postcondition of assignment
by dariomt＠gmail.com 26 Jun '09

26 Jun '09

Hi list, I see in the MultiArray concept documentation that the operator= has the postcondition (*.this) == x; But if I use overlapping memory this postcondition does not hold. The attached program asserts trying to verify the postcondition. In this program I have a 2D square matrix represented as a multi_array. I then assign column j to row i. Depending on the relationship of i and j and how the implementation of operator= does the traversal over the elements to be copied, the program may not satisfy the mentioned postcondition. May I suggest some note in the documentation about this. Is it possible that the traversal can be done forward or backward to guarantee the said postcondition even in the face of overlapping memory? Thanks. PS: Maybe I'm just using multi_array incorrectly. It that's the case, please do help me to use it correctly. #include <boost/multi_array.hpp> #include <iostream> #include <iterator> typedef boost::multi_array<double, 2> array_type; array_type::array_view<1>::type row(array_type & m, size_t i) { typedef array_type::index_range range; array_type::index_gen indices; return m[ indices[i][range(0,m.shape()[1])] ]; } array_type::array_view<1>::type col(array_type & m, size_t i) { typedef array_type::index_range range; array_type::index_gen indices; return m[ indices[range(0,m.shape()[0])][i] ]; } void print(array_type & m) { typedef array_type::index_range range; array_type::index_gen indices; for (size_t i = 0; i < m.size(); ++i) { array_type::array_view<1>::type r = row(m, i); std::copy( r.begin(), r.end(), std::ostream_iterator<double>(std::cout, " ")); std::cout << std::endl; } std::cout << std::endl; } int main() { typedef array_type::index index; array_type A(boost::extents[3][3]); int values = 1; for(index i = 0; i < 3; ++i) for(index j = 0; j < 3; ++j) A[i][j] = values++; print(A); array_type::array_view<1>::type r = row(A, 2); array_type::array_view<1>::type c = col(A, 1); c = r; print(A); // this assert fires because the contents of the row // and the column are not equal after the assigment // the postcondition does not hold assert( c == r ); return 0; }

3 2

[Flyweight] Enumerating all values in the factory?
by mailinglists 26 Jun '09

26 Jun '09

Hi, thanks for the flyweight library, looks good. Is there a way to enumerate all values in a factory? Consider my use case: I have a large set of data. Each element has a user defined 'tag', implemented as a flyweight<std::string>. Now the user wants to find all elements where the 'tag' matches a regex. Calculating the regex match for all elements in the dataset is expensive. If I could find all matching tags in the flyweight factory first, searching the elements is reduced to comparing two flyweights rather than doing a regex match. Consider: struct data { flyweight<std::string> m_tag; }; vector<data> m_some_data; The straight forward approach is costly: for (iterator i=m_some_data.begin(); i!=m_some_data.end() && regex_match(*i); ++i) ; with the regex match calculated for every element in the vector. A much more efficient solution is to find all elements in the flyweight factory that match the regex and then compare the flyweights. vector<flyweight<std::string> > m_matching_tags=find_matching_flyweights(regex); Now iterating over the data needs to compare two flyweights per data element which should be a lot faster than doing a regex match. Is there a way? Regards Hajo

2 1

[MultiArray] Get resize + copy in multi_array::operator=
by dariomt＠gmail.com 26 Jun '09

26 Jun '09

Hi list, What is the recommended way to get resize + copy behavior in multi_array's operator=? Should I patch boost source, wrap a multi_array object, inherit from multi_array? I guess I'm not the only one willing to get the usual semantics from the operator= of this container. TIA

1 0

How to implement ping using boost::asio::ip::icmp
by michael trudgian 26 Jun '09

26 Jun '09

Hoping someone can help me. I am trying to implement a basic ICMP ping using boost::asio. The boost asio library documentation concerning ICMP is pretty scant. It illustrates how to connect an icmp socket, but not how to send an echo request packet, or any other details about how the protocol can be used. If anyone has successfully used boost::asio::ip::icmp, and has some advice on how to use it to implement a simple ICMP ping, I would be greatly appreciative. _________________________________________________________________ POP access for Hotmail is here! Click here to find out more http://windowslive.ninemsn.com.au/article.aspx?id=802246

1 0

problem with boost::program_options - unregistered option in Config File
by Hasanul Ferdaus 25 Jun '09

25 Jun '09

Thanks for your reply. I am using boost::program_options to parse a config file like below --- sample.conf --- node.initiator=true tcp.port=8080 bootstrap.hints={ip{127.0.0.1},tcp(ip,{5002})} ---------------------------- Below I give the code that tries to read the config file sample.conf and tries to get all the options mentioned in file sample.conf - ------- Code ----- void main(){ std::string configFileName("sample.conf"); boost::program_options::options_description *optionsDescription; boost::program_options::variables_map *variablesMap; optionsDescription = new boost::program_options::options_description("Configuration"); variablesMap = new boost::program_options::variables_map(); optionsDescription->add_options() ("tcp.port", "TCP Port Number") ("bootstrap.hints", "Bootstrap Hints"); std::ifstream ifs(configFileName.c_str()); store(parse_config_file(ifs, *optionsDescription, true), *variablesMap); notify(*variablesMap); if (variablesMap->count("node.initiator"){ std::cout <<(*variablesMap)["node.initiator"].as<std::string>() << "\n"; } if (variablesMap->count("tcp.port")){ std::cout <<(*variablesMap)["tcp.port"].as<std::string>() << "\n"; } if (variablesMap->count("bootstrap.hints")){ std::cout <<(*variablesMap)["bootstrap.hints"].as<std::string>() << "\n"; } delete optionsDescription; delete variablesMap; return ; } ----------------- Now, the output is like below - --- output --- 8080 {ip{127.0.0.1},tcp(ip,{5002})} --------- But the value of the option node.initiator (i.e. "true") is not shown, also there is no exception. If I pass "false" in the 3rd parameter of parse_config_file() rather than "true", it throws exception, as it should do. But my question is, why "node.initiator" option is not in the variablesMap, as parse_config_file() should have parsed both registered and unregistered options. Surely if(variablesMap->count("node.initiator") does not pass. In the following link, it is mentioned that "allow_unregistered = false" is the default argument for parse_config_file() function, but I am surely passing "true" in my code. http://www.boost.org/doc/libs/1_38_0/doc/html/boost/program_options/parse_c… In another post you mentioned that SVN HEAD supports unregistered options in config file. Just now I have build the program_option library from boost svn, but there is no difference in the result. Awaiting your response. Thank you. --- On Thu, 6/25/09, Hasanul Ferdaus <sunny_register(a)yahoo.com> wrote: From: Hasanul Ferdaus <sunny_register(a)yahoo.com> Subject: problem with boost::program_options To: boost-users(a)lists.boost.org Date: Thursday, June 25, 2009, 9:41 AM Hi, I am trying to parse unregistered options using boost::program_options for config files but it does not recognize the unregistered options. the code is like to the following - -------------------------------- boost::program_options::options_description *optionsDescription; boost::program_options::variables_map *variablesMap; optionsDescription->add_options() ("ariba.bootstrap.hints", "Bootstrap Hints"); std::ifstream ifs(configFileName.c_str()); store(parse_config_file(ifs, *optionsDescription, true), *variablesMap); notify(*variablesMap); ---------------------------------- As it says, passing true to the 3rd argument of parse_config_file() should allow unregistered options to be parsed, but it does not recognize any unregistered options in config file. If anyone knows how to enable unregistered options in config file, please respond. Thanks in advance. Sunny

2 1

Re: [Boost-users] [Boost.Test] [1.38] Automated test suie registration with muti-file-test-modules
by Greg Christopher 25 Jun '09

25 Jun '09

> > > > Hi, > > We are starting to expand the number of test modules and are having > > some growing pains. > > > > Originally we were using the "single header" variant > > included/unit_test.hpp file. Now we have switched > to linking to the > dynamic version of the boost test library > and are including > auto_unit_test.hpp. > > Just unit_test.hpp is fine. Older name is deprecated. Ok- hopefully that's true for the version mentioned in the subject line. > > > We have been counting on the Test Tools to automatically register > > suites and test cases. This still worked when we just included > > unit_test.hpp (instead of included/unit_test.hpp). > > > > However with multiple test modules, init_unit_test_suite is being > > defined in each module. > > This should work as you expect. Please post simple example > illustrating your issues. Test1.cpp: ----- #include "stdafx.h" #define BOOST_TEST_MODULE foo2 #include <boost/test/execution_monitor.hpp> #include <boost/test/unit_test.hpp> class DefaultFixture { public: DefaultFixture() { } ~DefaultFixture() {}; }; BOOST_AUTO_TEST_SUITE(foo2) BOOST_FIXTURE_TEST_CASE(foo2, DefaultFixture) { BOOST_CHECK(4 == 4) } ----- Test2.cpp: ----- #include "stdafx.h" #define BOOST_TEST_MODULE foo #include <boost/test/execution_monitor.hpp> #include <boost/test/unit_test.hpp> class DefaultFixture { public: DefaultFixture() { } ~DefaultFixture() {}; }; BOOST_AUTO_TEST_SUITE(foo) BOOST_FIXTURE_TEST_CASE(foo, DefaultFixture) { BOOST_CHECK(4 == 4); } BOOST_AUTO_TEST_SUITE_END() ----- And Link with boost library.... 1>TestCase2.obj : error LNK2005: "class boost::unit_test::test_suite * 1>__cdecl init_unit_test_suite(int,char * * const)" 1>(?init_unit_test_suite@@YAPAVtest_suite@unit_test@boost@@HQAPAD@Z) 1>already defined in basic.obj Googling around, I discovered this cryptic define done in one of the files will fix it: ----- #define init_unit_test_suite init_auto_unit_test_suite ----- A strange observation as well.... Take this code: #include "stdafx.h" #define BOOST_TEST_MODULE foo #include <boost/test/execution_monitor.hpp> #include <boost/test/unit_test.hpp> class DefaultFixture { public: DefaultFixture() { } ~DefaultFixture() {}; }; BOOST_AUTO_TEST_SUITE(foo) BOOST_FIXTURE_TEST_CASE(foo, DefaultFixture) { BOOST_CHECK(4 == 4); } BOOST_AUTO_TEST_SUITE_END() And change BOOST_FIXTURE_TEST_CASE toBOOST_TEST_CASE(foo)... it won't compile: C:\basic\test2.cpp(21) : error C2882: 'foo' : illegal use of namespace identifier in expression c:\code\main\test\bora-vmsoft\svga\tests\bat\basic\basic\test2.cpp(22) : error C2448: 'boost::unit_test::make_test_case' : function-style initializer appears to be a function definition ----- Anyway, thanks for your help. The documentation right now seems to be lacking a little bit in giving clear examples of the different styles in which we are able to do test and suite registration. Greg

1 0

changing file output suffixes
by McLemon, Sean 25 Jun '09

25 Jun '09

Hi, Our toolchain generates files ending in ".doj" for objects, so I've been trying to configure my custom toolset module to use this. My original module is based on acc.jam but qcc.jam had something that looked like it would do the trick, so I lifted that and chucked into my module: import type ; type.set-generated-target-suffix OBJ : <toolset>bfin : doj ; type.set-generated-target-suffix STATIC_LIB : <toolset>bfin : dlb ; However my build of Boost now fails, with a fairly long error tracing: smclemo@edin-angus /usr/src/boost_1_39_0 $ bjam toolset=bfin WARNING: No python installation configured and autoconfiguration failed. See http://www.boost.org/libs/python/doc/building.html for configuration instructions or pass --without-python to suppress this message and silently skip all Boost.Python targets Building C++ Boost. After the build, the headers will be located at C:\cygwin\usr\src\boost_1_39_0 The libraries will be located at C:\cygwin\usr\src\boost_1_39_0\stage\lib Use 'bjam install --prefix=<path>' if you wish to install headers and libraries to a different location and remove the source tree. C:/cygwin/usr/src/boost_1_39_0/tools/build/v2/build\property.jam:613: in find-replace from module object(property-map)@1 error: Ambiguous key <target-type>OBJ <asynch-exceptions>off <conditional>@Jamfile</C:/cygwin/usr/src/boost_1_39_0>%Jamfile</C:/cygwi n/usr/src/boost_1_39_0>.handle-static-runtime <debug-symbols>on <define>BOOST_ALL_NO_LIB=1 <define>BOOST_DATE_TIME_STATIC_LINK <define>DATE_TIME_INLINE <exception-handling>on <extern-c-nothrow>off <hardcode-dll-paths>true <host-os>windows <include>. <inlining>off <install-dependencies>off <link>static <main-target-type>LIB <optimization>off <os>NT <preserve-test-targets>on <profiling>off <python-debugging>off <python>2.5 <rtti>on <runtime-debugging>on <runtime-link>shared <stdlib>native <suppress-import-lib>false <symlink-location>project-relative <tag>@Jamfile</C:/cygwin/usr/src/boost_1_39_0>%Jamfile</C:/cygwin/usr/sr c/boost_1_39_0>.tag <target-os>windows <target>object(file-target)@429 <threadapi>win32 <threading>multi <toolset-bfin:version>8.0.7.1 <toolset>bfin <user-interface>console <variant>debug <warnings-as-errors>off <warnings>on <snip> I'm guessing I'm doing something probably a bit stupid or missing something (I'm still at the fiddling around stage yet, so this is very likely). Is anyone in the position to confirm this, or do I need to send the module around? Also is it OK that I'm posting this here, or is there a more appropriate mailing list I should be using? - Sean

2 1

[ptr_container]delete_clone and shared libraries
by Robert Dailey 25 Jun '09

25 Jun '09

On Windows, I plan to create a ptr_vector in a separate DLL and propagate it to the executable for ownership (Factory-style). The executable will be responsible for the lifetime of this ptr_vector, and will eventually result in "delete" being called on each element in the ptr_vector. However, this 'delete' will be from the executable's memory manager, when the memory was originally allocated in the shared library's memory manager, causing serious issues. It would be ideal to specify a delete_clone class that ptr_vector would use (via ADL) and would force those objects in the ptr_vector to use the 'delete' call from the DLL's memory manager, and not the executable's. I don't see that this is going to be possible, or at least able to be done easily. Since the delete_clone's operator will have to be implemented in a CPP file, so it will be included into the shared library. This also implies that I'll have to import the delete_clone's interface, since this DLL will be loaded dynamically with LoadLibrary() in Windows. I just don't see this happening. Now, if ptr_vector had a function like so: ptr_vector::set_deleter() I could give it a pointer to a global function in my DLL, and the deleter becomes fully transparent to my executable. It just has to keep the ptr_vector around and let it fall out of scope and the deleter from the DLL will be used to delete each object in the ptr_vector. However, it isn't this simple and I need advice on how I can make this work. Thanks.

2 1